View FDR signficant and replicated findings across sub-Saharan Africa

Number of variables found in ‘X’ number of surveys

NUM_SURVEY_THRESHOLD <- 25
variables_num_surveys <- as.tibble(table(meta_analysis_continent$num_surveys)) %>% rename(num_surveys=Var1)
variables_num_surveys$num_surveys <- as.numeric(variables_num_surveys$num_surveys)
DT::datatable(variables_num_surveys, rownames = F)
NUM_SURVEY_THRESHOLD
## [1] 25
sum((variables_num_surveys %>% filter(num_surveys >= NUM_SURVEY_THRESHOLD))[, 'n'])
## [1] 88

plot the variables found in >= X surveys

meta_analysis_continent <- merge(varDesc, meta_analysis_continent)
meta_analysis_high_cnt <- meta_analysis_continent %>% filter(num_surveys >= NUM_SURVEY_THRESHOLD)

for(ii in 1:nrow(meta_analysis_high_cnt)) {
  mod <- (meta_analysis_high_cnt[ii, 'meta'])
  varname <- meta_analysis_high_cnt[ii, 'var']
  varlabel <- meta_analysis_high_cnt[ii, 'varlabel']
  gender <- meta_analysis_high_cnt[ii, 'gender']
  i2 <- mod[[1]]$I2
  forest(mod[[1]])
  title(sprintf('%s: %s (%s) I2=%02f', gender, varlabel, varname, i2))
}